metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-Instruct-v0.1
tags:
- trl
- sft
- generated_from_trainer
datasets:
- super_glue
metrics:
- accuracy
model-index:
- name: original_glue_boolq
results: []
original_glue_boolq
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on the super_glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.3297
- Accuracy: 0.8700
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 2
- eval_batch_size: 4
- seed: 2
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.4632 | 0.05 | 50 | 0.4840 | 0.7958 |
| 0.3453 | 0.1 | 100 | 0.3888 | 0.8226 |
| 0.2722 | 0.15 | 150 | 0.3590 | 0.8396 |
| 0.3266 | 0.2 | 200 | 0.3811 | 0.8459 |
| 0.3699 | 0.25 | 250 | 0.3534 | 0.8438 |
| 0.3554 | 0.3 | 300 | 0.3378 | 0.8565 |
| 0.1229 | 0.35 | 350 | 0.3368 | 0.8643 |
| 0.3522 | 0.4 | 400 | 0.3424 | 0.8643 |
| 0.2548 | 0.45 | 450 | 0.3467 | 0.8664 |
| 0.2119 | 0.5 | 500 | 0.3439 | 0.8714 |
| 0.2113 | 0.55 | 550 | 0.3518 | 0.8657 |
| 0.2122 | 0.6 | 600 | 0.3110 | 0.8770 |
| 0.3251 | 0.65 | 650 | 0.3323 | 0.8728 |
| 0.2904 | 0.7 | 700 | 0.3152 | 0.8792 |
| 0.6366 | 0.75 | 750 | 0.3502 | 0.8763 |
| 0.4161 | 0.8 | 800 | 0.3250 | 0.8806 |
| 0.1605 | 0.85 | 850 | 0.3258 | 0.8834 |
| 0.271 | 0.9 | 900 | 0.3330 | 0.8848 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.1+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0